Create main.py

main.py

@@ -0,0 +1,23 @@
+from ctransformers import AutoModelForCausalLM
+from fastapi import FastAPI, Form
+from pydantic import BaseModel
+
+#Model loading
+llm = AutoModelForCausalLM.from_pretrained("TheBloke/Toppy-M-7B-GGUF", model_file="toppy-m-7b.Q5_K_S.gguf", model_type="mistral", gpu_layers=50)
+
+
+#Pydantic object
+class validation(BaseModel):
+    prompt: str
+    
+#Fast API
+app = FastAPI()
+
+#Zephyr completion
+@app.post("/llm_on_gpu")
+async def stream(item: validation):
+    system_prompt = 'Below is an instruction that describes a task. Write a response that appropriately completes the request.'
+    E_INST = "</s>"
+    user, assistant = "<|user|>", "<|assistant|>"
+    prompt = f"{system_prompt}{E_INST}\n{user}\n{item.prompt.strip()}{E_INST}\n{assistant}\n"
+    return llm(prompt)